DEV Community

Cover image for The Complete Cypher Cheat Sheet
Memgraph profile image Memgraph
Memgraph for Memgraph

Posted on • Originally published at memgraph.com

The Complete Cypher Cheat Sheet

#cypher #memgraph #database #graphdatabase

Introduction

Cypher is the most widely adopted, fully specified, and open query language for property graph databases. It provides an intuitive and fast way to work with property graphs.

This article contains some of the most useful and common Cypher queries and their explanations. Whenever you are not sure how to write a Cypher query, you can take a look at this cheat sheet and try again. If you are new to graph databases and Cypher, you can also use this post to get acquainted with some of the features that Cypher and Memgraph offer.

1. Matching

Find nodes with specific properties 



MATCH (c:City)
WHERE c.name = "London"
RETURN c.population_size;
Enter fullscreen mode Exit fullscreen mode
  • MATCH (c:City): the MATCH clause specifies a node pattern with the label City and assigns the matches to variable c.
  • WHERE c.name = "London": the WHERE clause filters the matched results to those with a name property with value London.
  • RETURN c.population_size: the RETURN clause is used to request specific results.

This query can be rewritten without a WHERE clause to achieve the same result.



MATCH (c:City {name: "London"})
RETURN c.population_size;
Enter fullscreen mode Exit fullscreen mode

Find nodes with specific relationships 



MATCH (city:City)-[:IN]-(country:Country)
WHERE city.name = "London"
RETURN country.name;
Enter fullscreen mode Exit fullscreen mode
  • MATCH (city:City)-[:IN]-(country:Country): the MATCH clause specifies a node and relationship pattern with two connected nodes, labeled City and Country, connected by a relationship of type IN.

Match labels 



MATCH (c:City)
RETURN c;
Enter fullscreen mode Exit fullscreen mode

This query can be rewritten using the WHERE clause to achieve the same result.



MATCH (c)
WHERE c:City
RETURN c;
Enter fullscreen mode Exit fullscreen mode

Match multiple labels 



MATCH (c:City:Country)
RETURN c;
Enter fullscreen mode Exit fullscreen mode

This query can be rewritten using the WHERE clause to achieve the same result.



MATCH (c)
WHERE c:City AND c:Country
RETURN c;
Enter fullscreen mode Exit fullscreen mode

Matching nodes with properties in a range 



MATCH (c:City)
WHERE c.population_size >= 1000000 AND c.population_size <= 2000000
RETURN c;
Enter fullscreen mode Exit fullscreen mode

2. Creating

Create a node 



CREATE (c:City {name: "Zagreb", population_size: 1000000});
Enter fullscreen mode Exit fullscreen mode
  • c:City: creates a new node with the label City and assigns it to variable c (which can be omitted if it's not needed).
  • {name: "Zagreb", population_size: 1000000}: the newly created node has two properties, one with a string value and another with an integer value.

Create nodes with relationships 



CREATE (c1:City {name: "UK"}),
       (c2:City {name: "London", population_size: 9000000})
       (c1)<-[r:IN]-(c2)
RETURN c1, c2, r;
Enter fullscreen mode Exit fullscreen mode

The CREATE clause is used to create two new nodes and a directed relationship between them.

Create a relationship between existing nodes 



MATCH (c1), (c2)
WHERE c1.name = "UK" AND c2.name = "London"
CREATE (c2)-[:IN]->(c1);
Enter fullscreen mode Exit fullscreen mode

This will create a directed relationship of type IN between two existing nodes. If such a relationship already exists, this query will result in a duplicate. To avoid this, you can use the MERGE clause:



MATCH (c1), (c2)
WHERE c1.name = "UK" AND c2.name = "London"
MERGE (c2)-[:IN]->(c1);
Enter fullscreen mode Exit fullscreen mode

3. Updating

Add or update node properties 



MATCH (c:Country {name: "UK"})
SET c.name = "United Kingdom";
Enter fullscreen mode Exit fullscreen mode

If you use the SET clause on a property that doesn't exist, it will be created.

Replace all node properties 



MATCH (c:Country)
WHERE c.name = "United Kingdom"
SET c = {name: "UK", population_size: "66650000"};
Enter fullscreen mode Exit fullscreen mode
  • SET c = {name: "UK" ...}: this SET clause will delete all existing properties and create the newly specified ones.

Update multiple node properties 



MATCH (c:Country)
WHERE c.name = "United Kingdom"
SET c += {name: "UK", population_size: "66650000"};
Enter fullscreen mode Exit fullscreen mode
  • SET c += {name: "UK" ...}: this SET clause will add new properties and update existing ones if they are already set.

Check if a property exists and update it 



MATCH (c:Country)
WHERE c.name = "Germany" AND c.language IS NULL
SET c.language = "German";
Enter fullscreen mode Exit fullscreen mode

Because the WHERE clause contains the statement c.language IS NULL, the node will only be matched if it doesn't have a language property.

Rename a property 



MATCH (c:Country)
WHERE c.official_language IS null
SET c.official_language = c.language
REMOVE c.language;
Enter fullscreen mode Exit fullscreen mode
  • WHERE c.official_language IS null: the WHERE clause makes sure that you only create the new property in nodes that don't have a property with the same name.
  • SET n.official_language = n.language: you are technically not renaming a property but rather creating a new one with a different name and the same value.
  • REMOVE n.language: the REMOVE clause is used to delete the old property.

4. Deleting

Delete a node 



MATCH (c)-[r]-()
WHERE c.name = "US"
DELETE r, c;
Enter fullscreen mode Exit fullscreen mode
  • DELETE r, c: before you can delete a node, all of its relationships must be deleted first.

This query can be rewritten with the DETACH clause to achieve the same result.



MATCH (c)
WHERE c.name = "US"
DETACH DELETE c;
Enter fullscreen mode Exit fullscreen mode

Delete a property 



MATCH (c:Country)
WHERE c.name = "US" AND c.language IS NOT null
DELETE c.language;
Enter fullscreen mode Exit fullscreen mode

This query will delete the property language from a specific node.

Delete label in every node 



MATCH (c)
DELETE c:Country;
Enter fullscreen mode Exit fullscreen mode

This query will delete the label Country from every node.

Delete one of multiple labels 



MATCH (c)
WHERE c:Country:City
REMOVE c:City;
Enter fullscreen mode Exit fullscreen mode

This will delete the label City from every node that has the labels Country and City.

Delete all nodes and relationships 



MATCH (n)
DETACH DELETE n;
Enter fullscreen mode Exit fullscreen mode

This query will delete the whole database.

5. Constraints

Create a uniqueness constraint 



CREATE CONSTRAINT ON (c:City)
ASSERT c.location IS UNIQUE;
Enter fullscreen mode Exit fullscreen mode

This query will make sure that every node with the label City has a unique value for the location property.

Create an existence constraint 



CREATE CONSTRAINT ON (c:City)
ASSERT exists (c.name);
Enter fullscreen mode Exit fullscreen mode

This query will make sure that every node with the label City has the property name.

Check constraints 



SHOW CONSTRAINT INFO;
Enter fullscreen mode Exit fullscreen mode

This query will list all active constraints in the database.

Drop a uniqueness constraint 



DROP CONSTRAINT ON (c:City)
ASSERT c.location IS UNIQUE;
Enter fullscreen mode Exit fullscreen mode

This query will remove the specified uniqueness constraint.

Drop an existence constraint 



DROP CONSTRAINT ON (c:City)
ASSERT exists (c.name);
Enter fullscreen mode Exit fullscreen mode

This query will remove the specified existence constraint.

6. Graph Algorithms

To find out more about the built-in algorithms in Memgraph, take a look at the reference guide.

Breadth First Search 



MATCH (c1:City {name: "London"})-[
        edge_list:ROAD_TO *bfs..10
    ]-(c2:City {name: "Paris"})
RETURN *;
Enter fullscreen mode Exit fullscreen mode

This query will find all paths of length up to 10 between nodes c1 and c2.

Weighted Shortest Path 



MATCH (c1:City {name: "London"})-[
        edge_list:ROAD_TO *wShortest 10 (e, n | e.weight) total_weight
    ]-(c2:City {name: "Paris"})
RETURN *;
Enter fullscreen mode Exit fullscreen mode

The above query will find the shortest path of length up to 10 nodes between nodes c1 and c2. The length restriction parameter is optional.

7. NetworkX

If you want to know which NetworkX algorithms are available in Memgraph, take a look at the reference guide.

Analyze the whole graph 



CALL graph_analyzer.analyze() WITH YIELD *;
Enter fullscreen mode Exit fullscreen mode

This query will return various information like the number of nodes, number of edges, average degree, etc.

Find weakly connected components (Union Find) 



MATCH (n)-[e]->()
WITH collect(n) AS nodes, collect(e) AS edges
CALL wcc.get_components(nodes, edges) YIELD *
RETURN n_components, components;
Enter fullscreen mode Exit fullscreen mode

This query will search the whole graph for weakly connected components.

Calculate PageRank for all nodes 



CALL nxalg.pagerank() YIELD *
RETURN node.name AS name, rank
ORDER BY rank DESC
LIMIT 10;
Enter fullscreen mode Exit fullscreen mode

This query will calculate the rank of every node, order them from highest to lowest and return the first 10 results.

8. Other Useful Cypher Queries

Count all nodes 



MATCH (n)
RETURN count(n);
Enter fullscreen mode Exit fullscreen mode

This query will return the number of nodes in the database.

Count all relationships 



MATCH ()-->()
RETURN count(*);
Enter fullscreen mode Exit fullscreen mode

This query will return the number of relationships in the database.

Limit the number of returned results 



MATCH (c:City)
RETURN c
LIMIT 5;
Enter fullscreen mode Exit fullscreen mode
  • LIMIT 5: this will limit the number of returned nodes to 5.

Specify an alias for results 



MATCH (c:Country)
WHERE c.name = "US"
RETURN c.population_size AS population
Enter fullscreen mode Exit fullscreen mode

By using AS with the RETURN clause, the property
population_size` will be returned with an alias.

Conclusion

Cypher is an extensive query language with lots of features, and this cheat sheet is a great starting point for mastering them. If you are new to Cypher, we suggest taking our ten-day Cypher email course.

You can also go through our Cypher lessons on Memgraph Playground,
an online platform for running Cypher queries on live data.

If you have any questions about Cypher that aren't covered in this article, please take a look at our Cypher manual. Feel free also to share your questions at our Discord server.

Read more about Cypher query language on memgraph.com

Top comments (0)

Read next

stringintech profile image

Optimizing PostgreSQL Mass Deletions with Table Partitioning

stringintech -

davkharbayar profile image

AI-Powered CV Analysis Platform

Davkharbayar -

fahminlb33 profile image

AI Systematic Literature Review with KawanPaper

Fahmi Noor Fiqri -

thesohailjafri profile image

5 Quick And Easy MongoDB Optimizations (part 1)

Sohail Jafri -